TNG: DeepSeek R1T Chimera

TNG: DeepSeek R1T Chimera, a merge of DeepSeek-R1 and DeepSeek-V3 (0324), demonstrates exceptional speed, consistently ranking among the fastest models with an Infinityth percentile across 8 benchmarks. It offers competitive pricing, placing in the 47th percentile across 7 benchmarks. The model exhibits outstanding reliability, achieving a 100% success rate across all 8 benchmarks, indicating minimal technical failures. In terms of performance across categories, DeepSeek R1T Chimera shows remarkable strengths in General Knowledge and Ethics, achieving perfect 100% accuracy in both, and is noted as the most accurate model at its price point and among models of similar speed. Its Coding capabilities are also strong, with 95.0% accuracy (97th percentile), and Reasoning is highly proficient at 98.0% accuracy (94th percentile). While its Hallucinations accuracy is 94.0%, indicating a good ability to acknowledge uncertainty, its Instruction Following performance is a notable weakness, with a 1.0% accuracy in one benchmark and 0.0% in another, suggesting significant challenges with complex multi-step instructions. Email Classification is solid at 98.0% accuracy. Overall, the model excels in knowledge-based and ethical reasoning tasks, but requires improvement in instruction adherence.